PG 1553+113: Periodic BL Lac & Binary Candidate

• PG 1553+113 is a high-synchrotron-peaked BL Lac object known for its featureless optical spectrum and periodic modulation (~2.1 years) across gamma-ray, optical, and UV bands.
• Multiwavelength observations support one-zone SSC models with steady VHE emissions, though discrepancies in Doppler factors between VLBI and SED studies highlight ongoing challenges.
• Evidence from jet precession, polarization variations, and correlated timing analyses fuels binary SMBH and jet-instability scenarios, making PG 1553+113 a prime laboratory for variability studies.

PG 1553+113, also known as 1ES 1553+113, is a BL Lac object and a high-synchrotron-peaked blazar whose non-thermal jet is closely aligned with the line of sight. It is detected from radio to very-high-energy gamma rays, shows a compact parsec-scale jet extending to the northeast, and has become a prime candidate for periodic blazar activity and for a supermassive black hole binary because multiwavelength monitoring has revealed recurrent modulation on a timescale of about $2.1$–$2.2$ years, while later X-ray and polarimetric studies exposed additional, possibly distinct dynamical components (Caproni et al., 2017, Abdollahi et al., 14 Jan 2025).

1. Classification, designation, and redshift

PG 1553+113 is classified throughout the literature as a BL Lac object, specifically an HBL or HSP blazar. Its optical spectrum is described as nearly featureless, and its synchrotron component peaks at high frequencies, in the UV/X-ray domain. In the optical, the jet continuum dominates over the host galaxy and nuclear line emission; at high energies, the source is a persistent GeV and TeV emitter (Orr, 2011, Collaboration et al., 2011).

The redshift remains a central unsettled parameter because no secure direct spectroscopic measurement from source-intrinsic lines is established in the cited works. Indirect constraints and adopted values differ across analyses. A VLBI precession study used $z = 0.49 \pm 0.04$, consistent with earlier bounds $0.39 \lesssim z \lesssim 0.58$ (Caproni et al., 2017). Optical variability and IXPE-based studies adopted $z = 0.433$ (Agarwal et al., 2020, Middei et al., 2023). VERITAS combined HST/COS and VHE arguments to bracket the source within $0.4 < z < 0.5$ (Orr, 2011). A photohadronic fit to 17 VHE spectra favored values near $z \simeq 0.51$–$0.56$, with 95 per cent confidence intervals of $0.500 < z < 0.537$ for the Saldana-Lopez et al. (2021) EBL model and $0.491 < z < 0.527$ for the Dominguez et al. (2011) model (Sahu et al., 28 Mar 2025). The redshift literature on PG 1553+113 therefore consists of indirect bounds and model-dependent estimates rather than a single universally adopted value.

2. Broadband emission and very-high-energy properties

The source has long served as a benchmark TeV BL Lac. MAGIC observations over 2005–2009 found a very soft and stable observed VHE spectrum with mean photon index $2.2$0, while the flux above 150 GeV varied modestly between 4% and 11% of the Crab Nebula flux (Collaboration et al., 2011). VERITAS later obtained a $2.2$1 detection over 2010–2011, measuring a 175–500 GeV power-law spectrum

$2.2$2

with $2.2$3, about 10% of the Crab flux (Orr, 2011).

A simultaneous March–April 2008 campaign combined optical, X-ray, HE, and VHE data from KVA, REM, RXTE/ASM, AGILE, and MAGIC. The average MAGIC spectrum in that campaign was

$2.2$4

and the contemporaneous broad-band SED showed the standard two-hump structure (0907.0740). Time-averaged SEDs have been modeled successfully with one-zone SSC scenarios. One representative MAGIC fit used $2.2$5 cm, $2.2$6, $2.2$7 G, $2.2$8, $2.2$9, and $z = 0.49 \pm 0.04$0, with $z = 0.49 \pm 0.04$1, $z = 0.49 \pm 0.04$2, $z = 0.49 \pm 0.04$3, and $z = 0.49 \pm 0.04$4 (Prandini et al., 2011). A 2008 simultaneous SSC fit used $z = 0.49 \pm 0.04$5 G, $z = 0.49 \pm 0.04$6 cm, and $z = 0.49 \pm 0.04$7 (0907.0740). These parameter sets differ because they describe different data selections and source states, but both support the classification of PG 1553+113 as an HBL/HSP source whose X-rays belong to the synchrotron component and whose GeV–TeV output is consistent with inverse-Compton emission.

3. Quasi-periodic variability across wavebands

The modern timing literature on PG 1553+113 began with the report of an apparent gamma-ray quasi-periodicity in Fermi-LAT data. Using 6.9 years of observations, the source showed a gamma-ray period

$z = 0.49 \pm 0.04$8

while the optical light curve yielded $z = 0.49 \pm 0.04$9 days $0.39 \lesssim z \lesssim 0.58$0 years, and the 15 GHz OVRO light curve showed a less regular oscillation near $0.39 \lesssim z \lesssim 0.58$1 years (Collaboration et al., 2015). Tavani et al. then extended the Fermi baseline to about 9 years, recovered five main gamma-ray cycles, and identified weaker “twin peaks” around several of the main GeV peaks, with supporting X-ray activity (Tavani et al., 2018).

A later Fermi-LAT reanalysis over 2008–2023 moved the source from tentative to substantially stronger periodic status. Using more than 15 years of LAT data, the gamma-ray modulation spans more than seven cycles and has an estimated significance of 4 sigma against stochastic red noise. Pulse timing over the full light curve gave

$0.39 \lesssim z \lesssim 0.58$2

with the old and new halves of the LAT data remaining phase-consistent at chance probability $0.39 \lesssim z \lesssim 0.58$3 (Abdollahi et al., 14 Jan 2025). In that study, the optical light curves are clearly correlated with the gamma-ray modulation, while the radio correlation may involve a nonconstant lag. The discrete cross-correlation analysis found the optical consistent with zero lag relative to the gamma rays, $0.39 \lesssim z \lesssim 0.58$4 days, whereas the radio lagged by $0.39 \lesssim z \lesssim 0.58$5 days over the full light curve and by $0.39 \lesssim z \lesssim 0.58$6 days in a later segment (Abdollahi et al., 14 Jan 2025).

Optical timing analyses independently reinforce the $0.39 \lesssim z \lesssim 0.58$7-year modulation. A 2005–2019 compilation of $0.39 \lesssim z \lesssim 0.58$8- and $0.39 \lesssim z \lesssim 0.58$9-band data yielded a median period of $z = 0.433$0 years and additionally found a secondary period of about 210 days after correction for long-term variations (Agarwal et al., 2020). By contrast, the X-ray band remains more ambiguous. A two-jet precession analysis of 2012–2020 X-ray data argued that the 2.2-year modulation is consistent with the recurrence of main X-ray flares and that weak gamma-ray rebrightenings often correspond to stronger X-ray brightenings (Huang et al., 2021). A later Swift-based timing study, using more than 10 years of X-ray monitoring, instead found a preferred X-ray modulation near $z = 0.433$1 years, about 35% shorter than the gamma-ray period (Aniello et al., 2024). The multiwavelength timing picture is therefore coherent in gamma rays, optical, and UV, less regular in radio, and still unsettled in X-rays.

4. Parsec-scale jet, kinematics, and polarization

Very-long-baseline radio work shows PG 1553+113 to be a compact core–jet source. At 15 GHz the jet extends to the northeast, and a CE-based Gaussian decomposition of 17 VLBI epochs from 2009 to 2016 identified seven moving components, C1–C7. Their proper motions span $z = 0.433$2, corresponding to apparent speeds of about $z = 0.433$3 to $z = 0.433$4, and their mean sky position angles range from $z = 0.433$5 to $z = 0.433$6. The component ejection epochs for C2–C7 coincide, within $z = 0.433$7, with peaks of flare-like gamma-ray features (Caproni et al., 2017).

A steady-precession fit to those component trajectories favored a counterclockwise solution with

$z = 0.433$8

and a source-frame precession period $z = 0.433$9 (Caproni et al., 2017). In that model the viewing angle oscillates from about $0.4 < z < 0.5$0 to $0.4 < z < 0.5$1, and the Doppler factor ranges from about 0.4 to 8.5. However, the maxima of the modeled Doppler factor are delayed relative to the main gamma-ray flares by about $0.4 < z < 0.5$2 years, indicating that the GeV peaks are not simply the times of maximum line-of-sight boosting (Caproni et al., 2017).

A different 2015–2017 VLBA campaign emphasized wobbling rather than a clean periodic radio response. In that study the jet position angle varied in the range $0.4 < z < 0.5$3–$0.4 < z < 0.5$4, the stacked residual image indicated a funnel opening angle $0.4 < z < 0.5$5, and no clear radio periodicity comparable to the gamma-ray modulation was recovered. The same work found an average rotation measure of about $0.4 < z < 0.5$6, an intrinsic brightness temperature of $0.4 < z < 0.5$7 K, and an estimated Doppler factor of $0.4 < z < 0.5$8 (Lico et al., 2020). This radio-side estimate is much smaller than the higher values often invoked in one-zone SED fits or in the precession model.

Polarimetry adds further stratification. Optical photo-polarimetry in 2009 showed significant inter-night changes in the linear polarization percentage and position angle, while the total optical flux remained nearly constant (Andruchow et al., 2011). IXPE later measured an X-ray polarization degree of $0.4 < z < 0.5$9 at $z \simeq 0.51$0, while during the same campaign the optical EVPA executed a smooth rotation of about $z \simeq 0.51$1 at a rate of $z \simeq 0.51$2 degrees per day and no similar rotation was seen in radio or X-rays (Middei et al., 2023). The published interpretation is that the X-ray and optical emitting regions are separate, or at most partially co-spatial, and that the X-ray synchrotron emission traces a shock-accelerated, energy-stratified electron population (Middei et al., 2023).

5. Binary-black-hole and precession scenarios

Several studies interpret PG 1553+113 within an SMBH-binary framework, but they differ in the proposed driver of the light-curve morphology. The VLBI precession paper argued that the $z \simeq 0.51$3-year jet precession can be driven by a secondary black hole torquing the primary accretion disk. In that framework, the gamma-ray phase offset can be explained either by disk-crossing events, analogous to OJ 287, or by inverse-Compton scattering of photons from the secondary-disk radiation field by electrons in the primary jet (Caproni et al., 2017). Quantitatively, requiring long-term gravitational-wave stability gave lower limits on the mass ratio of $z \simeq 0.51$4 for a disk with $z \simeq 0.51$5 and $z \simeq 0.51$6 for $z \simeq 0.51$7, with corresponding lower limits $z \simeq 0.51$8–0.59 yr and $z \simeq 0.51$9–$0.56$0 pc (Caproni et al., 2017).

Tavani et al. proposed a more explicitly dynamical binary with masses of order $0.56$1 and $0.56$2. In that picture, a main jet $0.56$3 launched by the heavier black hole is periodically stressed at periastron, triggering MHD–kinetic tearing instabilities, magnetic reconnection, and electron acceleration, while the “twin peaks” around the main gamma-ray maxima arise either from additional perturbations in $0.56$4 or from a second, precessing jet $0.56$5 launched by the lighter companion (Tavani et al., 2018).

An X-ray-motivated two-jet precession model instead assigns an active jet to each black hole. Fitting the 0.3–10 keV light curve with $0.56$6 years fixed, Hua et al. obtained $0.56$7, $0.56$8, $0.56$9, $0.500 < z < 0.537$0, $0.500 < z < 0.537$1, $0.500 < z < 0.537$2, $0.500 < z < 0.537$3, and $0.500 < z < 0.537$4, and from a jet–disk power relation inferred a primary black-hole mass $0.500 < z < 0.537$5, a secondary mass $0.500 < z < 0.537$6, and mass ratio $0.500 < z < 0.537$7 (Huang et al., 2021).

The most recent gamma-ray timing analysis also favors a binary explanation, but with a different emphasis. Rather than geometric precession alone, it argues that the $0.500 < z < 0.537$8-year gamma-ray clock is most naturally the orbital period of a sub-parsec binary in which orbital motion modulates the accretion flow feeding the jet. For total mass $0.500 < z < 0.537$9, the inferred separation is $0.491 < z < 0.527$0 pc, about $0.491 < z < 0.527$1, with mass ratio $0.491 < z < 0.527$2 and gravitational-wave inspiral times $0.491 < z < 0.527$3–$0.491 < z < 0.527$4 years (Abdollahi et al., 14 Jan 2025). In that formulation, the periodic signal is a pulsational accretion-flow phenomenon in a sub-parsec binary, while disk and jet precession, rotation, nutation, instabilities, or perturbations by massive stars or intermediate-mass black holes remain alternatives rather than the preferred explanation (Abdollahi et al., 14 Jan 2025).

6. Current picture and unresolved questions

Across the cited literature, PG 1553+113 emerges as a rare case in which several distinct lines of evidence point toward structured, long-timescale jet dynamics. The source is a persistent HBL/HSP TeV blazar with a soft observed VHE spectrum, successful SSC descriptions of its time-averaged SED, and a reproducible $0.491 < z < 0.527$5–$0.491 < z < 0.527$6-year modulation in gamma rays, optical, and UV (Collaboration et al., 2011, Abdollahi et al., 14 Jan 2025). Its parsec-scale radio morphology is compact, core dominated, and variable in direction, and its polarization behavior is clearly stratified between optical, X-ray, and radio emitting regions (Caproni et al., 2017, Middei et al., 2023).

At the same time, several central questions remain open. The redshift is still indirect and model-dependent. The radio band does not show a comparably clean periodicity, even though it is correlated with the gamma-ray cycle. X-ray timing analyses are not yet convergent: one line of work ties the X-ray flares to the 2.2-year modulation, while another finds a distinct $0.491 < z < 0.527$7-year timescale (Huang et al., 2021, Aniello et al., 2024). There is also a pronounced Doppler-factor tension: optical flare analyses discuss $0.491 < z < 0.527$8 from one-zone SED modeling, whereas VLBA kinematics yield $0.491 < z < 0.527$9, explicitly described as part of the “Doppler crisis” (Agarwal et al., 2020, Lico et al., 2020).

For that reason, the status of PG 1553+113 is best described as that of a prime candidate periodic blazar and prime candidate SMBH-binary host rather than as a uniquely solved system. The binary-driven precessing-jet picture is quantitatively developed and internally consistent in several independent studies, but alternatives remain active in the literature: accretion-flow instabilities, helical or wobbling jet geometries, Lense–Thirring-like effects, and perturbations by compact orbiting bodies are all explicitly discussed (Caproni et al., 2017, Abdollahi et al., 14 Jan 2025). What distinguishes PG 1553+113 is not that every parameter is settled, but that its timing, VLBI, polarization, and VHE properties together make it one of the most technically constrained laboratories for studying the intersection of blazar variability, jet dynamics, and candidate supermassive-black-hole binaries.